Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 4200 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.7868. Min Act: -0.0000

Data Index: 37875 (C4 (Web Text))

Max Activating Token Index: 1014

Text #1

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.8294. Min Act: -0.0000

Data Index: 471702 (C4 (Web Text))

Max Activating Token Index: 643

Text #2

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.8001. Min Act: -0.0000

Data Index: 125514 (C4 (Web Text))

Max Activating Token Index: 340

Text #3

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.8067. Min Act: -0.0000

Data Index: 1101338 (C4 (Web Text))

Max Activating Token Index: 353

Text #4

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.7386. Min Act: -0.0000

Data Index: 866616 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.7382. Min Act: -0.0000

Data Index: 389047 (C4 (Web Text))

Max Activating Token Index: 205

Text #6

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6965. Min Act: -0.0000

Data Index: 648053 (C4 (Web Text))

Max Activating Token Index: 761

Text #7

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6656. Min Act: 0.0000

Data Index: 241843 (C4 (Web Text))

Max Activating Token Index: 103

Text #8

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.7061. Min Act: -0.0000

Data Index: 1243173 (C4 (Web Text))

Max Activating Token Index: 774

Text #9

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6770. Min Act: 0.0000

Data Index: 1216803 (C4 (Web Text))

Max Activating Token Index: 54

Text #10

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6718. Min Act: -0.0000

Data Index: 89334 (C4 (Web Text))

Max Activating Token Index: 267

Text #11

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6576. Min Act: -0.0000

Data Index: 1158497 (C4 (Web Text))

Max Activating Token Index: 341

Text #12

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6265. Min Act: -0.0000

Data Index: 1241412 (C4 (Web Text))

Max Activating Token Index: 938

Text #13

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6411. Min Act: -0.0000

Data Index: 669455 (C4 (Web Text))

Max Activating Token Index: 765

Text #14

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6592. Min Act: -0.0000

Data Index: 1450156 (Python Code)

Max Activating Token Index: 505

Text #15

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6107. Min Act: -0.0000

Data Index: 1207208 (C4 (Web Text))

Max Activating Token Index: 120

Text #16

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6122. Min Act: 0.0000

Data Index: 1102745 (C4 (Web Text))

Max Activating Token Index: 863

Text #17

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6242. Min Act: -0.0000

Data Index: 596436 (C4 (Web Text))

Max Activating Token Index: 369

Text #18

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.6350. Min Act: -0.0000

Data Index: 1352874 (C4 (Web Text))

Max Activating Token Index: 384

Text #19

Max Range: 0.8294. Min Range: -0.8294

Max Act: 0.5983. Min Act: -0.0000

Data Index: 101328 (C4 (Web Text))

Max Activating Token Index: 645